Individualising breast cancer treatment to improve survival and minimise complications in older women: a research programme including the PLACE RCT.
BACKGROUND: Over 44,000 women are diagnosed with breast cancer annually in the UK. The research comprised three workstreams (WSs) focused on older women. MAXIMISING SURVIVAL: WS1 – to identify the role of older women’s and surgeons’ preferences in cancer treatment decisions and whether comorbidity or fitness for surgery has an impact on survival. MINIMISING COMPLICATIONS: WS2 – to assess multifrequency bioimpedance (BEA) compared with perometry in identifying women predisposed to develop lymphoedema after axillary node clearance (ANC) surgery. WS3 – to assess, in women at risk of lymphoedema, whether or not applying compression garments prevents the onset of lymphoedema. DESIGN: WS1 – a prospective, consecutive cohort of surgical consultations with women aged ≥ 70 years with operable breast cancer. Interviews and questionnaire surveys of surgeons’ and women’s perceptions of responsibility for treatment decisions (Controlled Preference Score), effects related to survival and secondary outcomes. WS2 – women undergoing ANC for cancer in 21 UK centres underwent baseline and subsequent BEA, and perometer arm measurements and quality-of-life (QoL) assessments. WS3 – a randomised controlled trial testing standard versus applying graduated compression garments to the affected arm, for 1 year, in WS2 patients developing arm swelling. SETTING: Breast outpatient clinics in hospitals with specialist lymphoedema clinics. PARTICIPANTS: WS1 – patients aged ≥ 70 years with newly diagnosed, operable, invasive breast cancer. WS2 – women with node-positive cancer scheduled to undergo ANC. WS3 – WS2 participants developing a 4–9% increase in arm volume. INTERVENTIONS: WS1 – observational study. WS2 – observational study. WS3 – application of graduated compression garments to affected arm, compared with standard management, for 1 year. OUTCOMES: WS1 – self-report and clinically assessed health, QoL, complications and survival. WS2 – perometer and bioimpedance spectroscopy (BIS) measurements, QoL and health utility; and sensitivity and specificity of BIS for detecting lymphoedema compared with perometer arm measurements; in addition, a health economics assessment was performed. WS3 – time to the development of lymphoedema [≥ 10% relative arm-volume increase (RAVI)] from randomisation. RESULTS: WS1 – overall, 910 women were recruited, but numbers in the substudies differ depending on consent/eligibility. In a study of patient/surgeon choice, 83.0% [95% confidence interval (CI) 80.4% to 85.6%] had surgery. Adjusting for health and choice, only women aged > 85 years had reduced odds of surgery [odds ratio (OR) 0.18, 95%CI 0.07 to 0.44]. Patient role in treatment decisions made no difference to receipt of surgery. A qualitative study of women who did not have surgery identified three groups: ‘patient declined’, ‘patient considered’ and ‘surgeon decided’. In a survival substudy, adjusting for tumour stage, comorbidity and functional status, women undergoing surgery had one-third the hazard of dying from cancer. Serious complications from surgery were low and not predicted by older age. In a substudy of the effect of surgical decision-making on HRQoL, 59 (26%) received preferred treatment decision-making style. In multivariate analyses, change in HRQoL was associated neither with congruence (p = 0.133) nor with receipt of surgery (p = 0.841). In a substudy of receipt of chemotherapy in women aged ≥ 65 years, adjusting for tumour characteristics, health measures and choice, women aged ≥ 75 years had reduced odds of chemotherapy (OR 0.06, 95%CI 0.02 to 0.16). WS2 – lymphoedema by 24 months was detected in 21.4% of women by perometry (24.4% sleeve application) and in 39.4% by BIS. Perometer and BIS measurements correlated at 6 months (r = 0.61). Specificity for sleeve application was greater for perometry (94% CI 93% to 96%) at 24 months, as was a positive predictive value of 59% (95% CI 48% to 68%). Lymphoedema diagnosis reduced QoL scores. Sleeve application in the absence of RAVI of > 9% did not improve QoL or symptoms. A composite definition of lymphoedema was developed, comprising a 9% cut-off point for perometer and self-reported considerable swelling. Diagnostic accuracy was ≥ 94% at 6, 12 and 24 months. WS3 – the PLACE (Prevention of Lymphoedema After Clearance of External compression) trial recruited 143 patients, but recruitment was slow and closed early on the advice of the Independent Data Monitoring Committee. A qualitative substudy identified a number of barriers to recruitment. CONCLUSIONS: Half of older patients felt that they influenced decisions about their treatment. No relationship between decision preference being fulfilled and HRQoL in elderly patients diagnosed with cancer occurred, and older age did not predict complications. Primary surgery reduced the hazard of dying of cancer by two-thirds, independent of age, health and tumour characteristics. Women aged ≥ 75 years have reduced odds of receiving chemotherapy. Lymphoedema (along with a BMI of > 30 kg/m2, cigarette smoking and chemotherapy) reduces QoL. Changes in arm volume of > 9% predicted lymphoedema requiring and benefiting from sleeve application. The PLACE trial qualitative work provides a number of insights into problems of recruitment that were specific to this trial (stigma of compression garments) but that are also generalisable to other RCTs. LIMITATIONS: Both WS1 and WS2 were large, multicentre, UK cohort, observational studies. The WS3 PLACE trial has not reported yet but closed with approximately half of the patients originally planned. FUTURE WORK: Research producing objective measures for sleeve prescription in the NHS is required. TRIAL REGISTRATION: Current Controlled Trials ISRCTN48880939. FUNDING: This project was funded by the National Institute for Health Research (NIHR) Programme Grants for Applied Research programme and will be published in full in Programme Grants for Applied Research; Vol. 7, No. 5. See the NIHR